AI-Enabled decision support systems for agrometeorological forecasting and climate change mitigation

Agriculture is challenged by climate change and can only be mitigated through in advanced forecasting and strategy. This research investigates AI enables decision support systems for agrometeorological forecasting and climate change mitigation. Four machine learning algorithms were then implemented to predict temperature, precipitation, and drought susceptibility: Random Forest, Long Short-Term Memory (LSTM), Support Vector Machines (SVM), Convolutional Neural Networks (CNN). With high predictive accuracy achieved, the models were trained using historical climate and agricultural data. The results showed that the CNN was outperformed by other models (with the highest achievement, the CNN, had the accuracy of 92.5%, and other models was LSTM (accuracy = 90.3%), Random Forest (accuracy = 88.7%) and SVM (accuracy = 85.2%). Where AI-based forecasting outperformed traditional forecasting methods was significantly reducing prediction errors by 23%. Proposed models also enhanced early disaster detection, reducing the agricultural losses 30%. Our approach was shown to be more adaptable as well as computationally efficient in climate forecasting than existing works. These findings emphasize how AI might better serve agricultural planning, resource allocation and climate adaptation planning. Future research should be on hybrid AI models and live monitoring system for global agricultural applications. However, the adverse effects of climate change on food production can be mitigated with an AI driven climate smart agriculture.